Cars.com
Marketplace AI building blocks monetized on both sides: conversational search on the buyer side for conversion, and generation and AI tools packaged and sold to subscribing dealers.
On November 6, 2025, Cars.com launched Carson, a multilingual natural-language AI search engine that assists about 15% of web and mobile web searches and improves conversion from results pages to vehicle detail pages by nearly 30%; AI serves the monetization of a marketplace where about 89% of the record $182 million Q3 2025 revenue comes from subscribing dealers, to whom Cars.com also sells AI building blocks such as inventory video (twice the lead conversion).
Key points
- Cars.com launched Carson, a multilingual natural-language AI search engine, on November 6, 2025.
- Carson assists about 15% of web and mobile web searches and lifts conversion from results pages to vehicle detail pages by about 30%.
- AI serves monetization: about 89% of the record $182 million Q3 2025 revenue comes from subscribing dealers.
- The AI building blocks are packaged and sold to dealers (AI-Powered Inventory Video, twice the lead conversion).
Objective
Turn AI into a monetization lever for an auto marketplace where about 89% of revenue comes from subscribing dealers. Cars.com uses the Carson engine on the buyer side to make its marketplace perform better (more engagement, better conversion from results to vehicle detail pages), which strengthens the value sold to dealers, then packages its AI building blocks into paid solutions offered to those same dealers to generate more leads and profit.
The deployment
Cars.com operates a US auto marketplace whose revenue comes mostly from subscribing dealers ($162 million out of $181.6 million in Q3 2025 revenue, or about 89%). On November 6, 2025, it launched Carson, a multilingual AI search engine that turns a natural-language query (for example a reliable, affordable vehicle for a family of five) into targeted results, in place of filters. More than 70% of visitors start their search with no make or model in mind, which Carson is meant to unlock. On the buyer side, Cars.com quantifies the effect: Carson assists about 15% of web and mobile web searches, its users return twice as often, save three times as many vehicles, and convert nearly 30% better from results pages to vehicle detail pages. CEO Alex Vetter summarized the impact on the Q3 2025 earnings call by pointing to a doubling of visitor engagement. On the monetization side, Cars.com turns these AI capabilities into products sold to its dealers: on January 29, 2026, it announced four solutions including AI-Powered Inventory Video, which generates vehicle-specific video ads and shows twice the lead conversion of standard video. The number of subscribing dealers reached 19,526 in Q3 2025, a three-year high, and dealer revenue returned to growth (up 2% year over year) in the record-revenue quarter.
Results Proof A
The monetization figures come from Exhibit 99.1 of the 8-K filed with the SEC for Q3 2025 (record revenue of $182 million, dealer revenue of $162 million, dealer growth of 2%, 19,526 dealers), an uninterpreted primary document. The Carson and dealer-solution metrics come from official brand press releases published on cars.com. Three concordant T1 sources, including a financial filing.
How it works
Inferred typical approachThe internal detail is not public. Here is a proven approach that leads to the same result, to adapt to your stack.
The stack in detail
- outil Carson Cars.com's multilingual AI search engine, launched November 6, 2025, which turns natural-language queries (for example a need for a reliable, affordable vehicle for a family) into targeted results, replacing filter-based navigation. Deployed on web and mobile web, with the mobile app rollout in pilot in Q4 2025.
- outil AI-Powered Inventory Video A solution that generates VIN-specific video ads with audience targeting, sold to dealers; presented as driving twice the lead conversion on their site compared to traditional video creation.
- outil AccuTrade Cars Commerce's appraisal and trade-in solution, integrated into the dealer offering; more than one million quarterly appraisals in Q3 2025.
How it runs, concretely
For ops teams-
1Buyer's natural-language query customer
The visitor describes their need in plain language (budget, use, reliability) instead of handling filters; more than 70% arrive with no make or model in mind.
-
2AI interprets and returns targeted results AI
Carson interprets the query and returns a tailored results page, with summaries (award-winning vehicles, made-in-America index), to drive the click toward vehicle detail pages.
-
3Conversion to the vehicle detail page and the dealer customer
The buyer moves from results to the vehicle detail page and then to the dealer; Cars.com measures SRP-to-VDP conversion, saves, and leads.
-
4Packaging the AI building blocks into sold solutions data team
Cars Commerce turns its AI capabilities into paid products (VIN-level inventory video, market-area expansion, reporting) offered to subscribing dealers.
-
5Dealer activation and subscription billing customer
The dealer subscribes and activates the AI solution on their inventory; the revenue flows into dealer revenue, which accounts for about 89% of the group's revenue.
Conversion from results pages to vehicle detail pages, and lead conversion on the dealer side. If the indexed inventory (listings, prices, photos) degrades, the relevance of Carson's answers and therefore conversion drop, and the value sold to dealers falls with it.
How your customers perceive this type of use
Sourced studiesLes consommateurs n'acceptent pas les chatbots par defaut : 64% prefereraient que les entreprises n'utilisent pas d'IA dans leur service client (Gartner, 2024) et pres d'un utilisateur sur cinq du service client par IA n'en retire aucun benefice (Qualtrics, 2025). L'acceptation se construit sur trois conditions mesurees par Salesforce : savoir qu'on parle a une IA, pouvoir escalader vers un humain, comprendre la logique de l'agent.
Acceptance conditions
- Etre informe qu'on parle a une IA et non a un humain (pres de 75% le demandent, Salesforce 2024)
- Un chemin d'escalade clair vers un agent humain (45% plus enclins a utiliser l'agent IA, Salesforce 2024)
- Une logique de l'agent clairement expliquee (44% plus enclins, Salesforce 2024)
Red lines
- Rendre l'humain injoignable : c'est la premiere inquietude des consommateurs sur l'IA dans le service client (Gartner 2024) et 50% craignent que l'IA les coupe du contact humain (Qualtrics 2025)
- Remplacer le service client par l'IA sans alternative : 53% envisageraient de partir chez un concurrent (Gartner 2024)
Sources: Salesforce 2024 · Gartner 2024 · Qualtrics 2025
How to replicate
Inference, not sourcedData prerequisites
- A rich, clean inventory catalog (listings, prices, photos, features) that can be indexed for natural-language search
- A first-party history of search behavior to frame the relevance of results
- Reliable measurement of conversion from results pages to product pages and to the lead, before and after
Org prerequisites
- A marketplace business model where most revenue comes from subscribing sellers, not from a margin on transactions
- A product and data team able to operate a conversational engine in production and to package AI building blocks into commercial offers
- A sales force able to sell and operate these AI solutions with sellers
- A compliance framework for AI transparency (AI Act) and the GDPR legal basis for personalization
Possible stack
- An in-house conversational search engine coupled with an LLM and an inventory index
- A per-listing video creative generator for sellers
- Market-area expansion and audience-targeting tools
- Consolidated self-service reporting for sellers
The plan, step by step
- Step 1Index the inventory (listings, prices, photos, attributes) to enable reliable, multilingual natural-language search.Deliverable: An inventory index queryable by a conversational assistant.
- Step 2Deploy a conversational search engine on the buyer side on a high-volume surface, with transparency about its AI nature.Deliverable: Engine in production, results traceable to the source inventory.
- Step 3Measure conversion from results pages to product pages and to the lead, on the equipped surface against the old path.Deliverable: Before-and-after conversion reading, decision to extend or stop.
- Step 4Package the AI capabilities into paid solutions for sellers (per-listing video generation, market-area expansion, reporting).Deliverable: A commercial offering of AI building blocks that subscribing sellers can activate.
- Step 5Track the effect on seller revenue and lead conversion to justify pricing and expand the solutions catalog.Deliverable: Reading of the incremental seller revenue attributable to the AI solutions.
First step: Connect a conversational search engine to the existing inventory of a single high-volume surface, and measure conversion from results pages to product pages against the old filter-based path, before turning it into a building block sold to sellers.
Sources
- S1 Meet Carson, Cars.com's New AI Engine for Car Shopping Primary archive pending
- S2 Cars.com Reports Third Quarter 2025 Results (Form 8-K, Exhibit 99.1) Primary archive pending
- S3 Cars.com Launches Powerful New Solutions Helping Dealers Turn Smarter Technology Into Real Profit Primary archive pending
An error, newer info, a source?
This page lives on its accuracy. If a figure has moved, if the deployment has changed, or if you have a higher-quality source, tell us. Every sourced correction is verified before publication.